def test_to_coord_string(self, molecule):
ms = MoleculeSystem(molecule)
test_value = """
0.00000000000000 0.00000000000000 0.00000000000000 c
0.00000000000000 0.00000000000000 2.86118897312869 o
0.00000000000000 0.00000000000000 -2.86118897312869 o
"""
dg = DataGroups(ms.to_coord_string())
assert len(dg.dg_list) == 2
check_dg(dg.sdg("coord", strict=True), test_value)
ms.frozen_indices = {0, 1}
test_value = """
0.00000000000000 0.00000000000000 0.00000000000000 c f
0.00000000000000 0.00000000000000 2.86118897312869 o f
0.00000000000000 0.00000000000000 -2.86118897312869 o
"""
dg = DataGroups(ms.to_coord_string())
assert len(dg.dg_list) == 2
check_dg(dg.sdg("coord", strict=True), test_value)
def test_dummy_atoms(self, molecule_filepath):
ms = MoleculeSystem.from_file(molecule_filepath, fmt="coord")
mol = ms.molecule
print(mol[-1].specie)
print(mol[-1].specie.__class__)
# Pymatgen's Specie and DummySpecie have been changed to Species and
# DummySpecies in v2020.10.9. We keep testing both for backward compatibility.
assert isinstance(mol[-1].specie, (DummySpecies, DummySpecie))
assert mol[-1].specie.symbol == "Q"
test_value2 = """
0.00000000000000 0.00000000000000 -0.12178983933899 o
1.41713420892173 0.00000000000000 0.96657854674257 h
-1.41713420892173 0.00000000000000 0.96657854674257 h
0.00000000000000 0.00000000000000 0.00000000000000 q
"""
dg = DataGroups(ms.to_coord_string())
assert len(dg.dg_list) == 2
check_dg(dg.sdg("coord", strict=True), test_value2)
assert_MSONable(ms)
def test_from_string(self):
# basic test
string = """
$coord
.00000000000000 .00000000000000 .00000000000000 n
-1.15103063747470 -1.99364354517457 .00000000000000 o
2.30206127494940 .00000000000000 .00000000000000 o
-1.15103063747470 1.99364354517457 .00000000000000 o
$end
"""
ms = MoleculeSystem.from_string(string=string, fmt="coord")
mol = ms.molecule
assert mol[1].coords[0] == pytest.approx(-0.6090991821345737)
assert len(mol) == 4
assert len(ms.frozen_indices) == 0
assert_MSONable(ms)
# no coord
with pytest.raises(ValueError,
match=r'^The string does not contain \$coord!$'):
MoleculeSystem.from_string(string="$end", fmt="coord")
# with frozen and internal definitions
string = """
$coord
.00000000000000 .00000000000000 .00000000000000 n f
-1.15103063747470 -1.99364354517457 .00000000000000 o
2.30206127494940 .00000000000000 .00000000000000 o
-1.15103063747470 1.99364354517457 .00000000000000 o f
$intdef
# definitions of internal coordinates
1 k 1.0000000000000 stre 1 2 val= 2.43987
2 f 1.0000000000000 bend 1 2 3
-0.5000000000000 bend 2 3 4
$end
"""
ms = MoleculeSystem.from_string(string=string, fmt="coord")
mol = ms.molecule
assert mol[1].coords[0] == pytest.approx(-0.6090991821345737)
assert len(mol) == 4
assert ms.frozen_indices == {0, 3}
assert len(ms.int_def) == 2
assert ms.int_def[0].value == pytest.approx(2.43987)
assert ms.int_def[1].value == None
assert ms.int_def[0].status == "k"
assert ms.int_def[1].status == "f"
assert ms.int_def[0].indices[0] == [0, 1]
assert len(ms.int_def[1].indices) == 2
assert ms.int_def[1].weights[1] == pytest.approx(-0.5)
dg = DataGroups(ms.to_coord_string())
dg_ref = DataGroups(string)
assert len(dg.dg_list) == 3
check_dg(dg.sdg("coord", strict=True), dg_ref.sdg("coord",
strict=True))
check_dg(dg.sdg("intdef", strict=True),
dg_ref.sdg("intdef", strict=True))
# with user-defined bonds
string = """
$coord
.00000000000000 .00000000000000 .00000000000000 n
-1.15103063747470 -1.99364354517457 .00000000000000 o
2.30206127494940 .00000000000000 .00000000000000 o
-1.15103063747470 1.99364354517457 .00000000000000 o
$user-defined bonds
1-2, 2 - 3,3|4
$end
"""
ms = MoleculeSystem.from_string(string=string, fmt="coord")
mol = ms.molecule
assert mol[1].coords[0] == pytest.approx(-0.6090991821345737)
assert len(mol) == 4
assert ms.user_defined_bonds == {(0, "-", 1), (1, "-", 2), (2, "|", 3)}
dg = DataGroups(ms.to_coord_string())
dg_ref = DataGroups(string)
assert len(dg.dg_list) == 3
check_dg(dg.sdg("coord", strict=True), dg_ref.sdg("coord",
strict=True))
check_user_defined_bonds_dg(
dg.sdg("user-defined bonds", strict=True),
dg_ref.sdg("user-defined bonds", strict=True))
# malformed user-defined bonds
string = """
$coord
.00000000000000 .00000000000000 .00000000000000 n
-1.15103063747470 -1.99364354517457 .00000000000000 o
2.30206127494940 .00000000000000 .00000000000000 o
-1.15103063747470 1.99364354517457 .00000000000000 o
$user-defined bonds
1-2, 2 3,3|4
$end
"""
with pytest.raises(ValueError,
match="Cannot parse user-defined bonds.*"):
MoleculeSystem.from_string(string=string, fmt="coord")
# from xyz format
ms = MoleculeSystem.from_string(mol.to(fmt="xyz"), fmt="xyz")
assert ms.molecule[1].coords[0] == pytest.approx(-0.6090991821345737)
def from_string(cls, string, fmt="coord"):
"""
Creates an instance from a string. Could be the string of a coord file
or any format supported by pymatgen Molecule.
Args:
string (str): the string with the data.
fmt (str): the format of the data. could be "coord" for Turbomole
coord file or any format supported in pymatgen Molecule.
Returns:
An instance of MoleculeSystem.
"""
if fmt == "coord":
dg = DataGroups(string=string)
coordinates_str = dg.sdg("$coord", strict=True)
if not coordinates_str:
raise ValueError("The string does not contain $coord!")
mol, fi = get_mol_and_indices_frozen(coordinates_str)
int_def_str = dg.sdg("$intdef", strict=True)
int_def = []
if int_def_str:
lines = []
# remove empty lines and comments
for l in int_def_str.splitlines():
lstrip = l.strip()
if lstrip and not lstrip.startswith("#"):
lines.append(l)
int_def_str = "\n".join(lines)
# split based on the presence of the index plus the status.
# In a case like this:
# 1 k 1.0000000000000 stre 4 1 val= 1.80084
# 2 k 1.0000000000000 bend 4 3 1 val= 106.27756
# 1.0000000000000 bend 3 2 1
# 1.0000000000000 bend 2 4 1
# 3 f 1.0000000000000 tors 1 2 3 4
# will split in 3 groups based on the presence of the digit plus k, f, d or i
# at the beginning of the line.
r = r"^\s*\d+\s+[kfdi]\s+.*?(?=\s*\d+\s+[kfdi]\s+|\Z)"
for group in re.findall(r, int_def_str,
re.DOTALL | re.MULTILINE):
int_def.append(InternalDefinition.from_string(group))
user_def_bonds_str = dg.sdg("$user-defined bonds", strict=True)
user_def_bonds = set()
if user_def_bonds_str:
# parses a line of this form:
# 1-2, 3-4, 5|6
# splitting first on "," and then on "-" and "|"
for l in user_def_bonds_str.splitlines():
l = l.strip()
if not l or l.startswith("#"):
continue
for bond in l.split(","):
for separator in ("-", "|"):
if separator in bond:
bond_indices = bond.split(separator)
if len(bond_indices) != 2:
raise ValueError(
"Cannot parse user-defined bonds for line: {}"
.format(l))
index_1 = int(bond_indices[0]) - 1
index_2 = int(bond_indices[1]) - 1
user_def_bonds.add(
(index_1, separator, index_2))
break
else:
raise ValueError(
"Cannot parse user-defined bonds for line: {}".
format(l))
return cls(mol,
int_def=int_def,
frozen_indices=fi,
user_defined_bonds=user_def_bonds)
else:
return cls(Molecule.from_str(string, fmt))
